Last tech talk Matt wanted to test the NETLOGO’s fishers behavior by seeing their reaction to MPAs. Matt argued that with MPAs set up you’ll see fishers congregate at the borders and fishing there. I wanted to test whether this is something our MASON model replicates without adding new fisher behavior.
Fishers proceed as always by exploration-imitation. They have no “knowledge” of MPAs and do not take them into account when making decisions. Whenever they explore a sea-patch that is protected they make 0 catches which “teaches” them that’s not such a hot location.
It turns out that even such simple agents do end up fishing just at the border of MPAs.
First I implemented a little gui functionality so that we can draw MPAs on the fly as the simulation is running:
The yellow box in the middle of the ocean is an MPA. Agents can still steam through it but they can’t fish on it.
Initially agents bypass the MPA, fishing elsewhere but as all the other sources of fish are exhausted a lot of fishers tend to fish just at the border of the protected zone, just like we’d like them to: I like this result very much because I didn’t hardcode anything in, it’s just a result of trial-and-error behavior by the agents themselves.
Although there is nothing telling agents to fish near MPAs the fact is that the only fish alive is there and the spreading of biomass over time means that there is an outflow of fish from the MPA to where fishermen can catch them.